Skip to content
Back to skills

Options Structure Analyzer

ASecurity

Given a view (direction_bullish, direction_bearish, vol_long, vol_short, hedge), a horizon, and a target move, enumerate candidate options structures (long call/put, bull/bear spreads, straddles, strangles, iron condor, protective put, collar), compute payoff-at-target, and rank by payoff/capital. Not a black-box recommendation — a structured comparison so the operator picks the structure whose tradeoffs match the view. Use when the operator has a directional or vol thesis and wants to see th...

  • 7 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 6, 2026
developmentgoexpressapi

Works with

  • api

Security analysis

A100/100

Pro scans all 5 files and shows the line behind each finding

Scanned September 6, 2026

npx -y skills add rgourley/quant-garage --skill options-structure-analyzer --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Options Structure Analyzer?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Options Structure Analyzer
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/rgourley-options-structure-analyzer/badge)](https://www.skillsdirectory.com/skills/rgourley-options-structure-analyzer)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: options-structure-analyzer
description: Given a view (direction_bullish, direction_bearish, vol_long, vol_short, hedge), a horizon, and a target move, enumerate candidate options structures (long call/put, bull/bear spreads, straddles, strangles, iron condor, protective put, collar), compute payoff-at-target, and rank by payoff/capital. Not a black-box recommendation — a structured comparison so the operator picks the structure whose tradeoffs match the view. Use when the operator has a directional or vol thesis and wants to see the options tradeoffs side by side.
---

# options-structure-analyzer

You hand over a ticker, a view (`direction_bullish`, `direction_bearish`,
`vol_long`, `vol_short`, or `hedge`), a horizon in days, and a target
move. The skill fetches the nearest expiry with priceable legs on
both sides, enumerates the candidate structures for your view,
computes payoff-at-target for each, and ranks them.

Not a recommendation. A structured comparison so you can pick the
structure whose tradeoffs match your thesis — capped-risk spread vs
unbounded-upside long, straddle vs strangle premium tradeoff, collar
credit vs protective-put cost.

## When to invoke

- The operator has a view and wants to know "how do I express this
  with options" — the answer depends on the tradeoff they prefer
- Comparing single-leg (long call) vs two-leg (spread) vs multi-leg
  (condor) structures before deciding
- Sizing hedges on an existing position (protective put vs collar)
- The user says "options structure", "how do I trade this with
  options", "should I buy the call or the spread"

## What you need

- `MASSIVE_API_KEY` with an options entitlement (Options Developer or
  higher). Chain snapshot is the primary data pull.

## What you get back

**Layer 1 JSON** matching [`output-schema.json`](./output-schema.json).
Per-structure block with legs (buy/sell + type + strike + ticker +
premium + qty), net debit/credit, max profit, max loss, breakevens,
capital required, and payoff at your target price.

**Layer 2 rendered comparison**. One block per structure with a
plain-English read, legs listed, key metrics, and payoff-at-target.
On hedge structures, the payoff line includes "vs unhedged" delta
rather than a meaningless percent-of-capital ratio. See
[`references/rendering.md`](./references/rendering.md).

## Views supported

- `direction_bullish` — long call, bull call spread
- `direction_bearish` — long put, bear put spread
- `vol_long` — long straddle, long strangle
- `vol_short` — short iron condor
- `hedge` — protective put, collar (assumes 100 shares long
  underlying)

## How it works

1. **Fetch spot** via the snapshot endpoint. Walks a strict fallback
   chain (lastTrade > min > day > prevDay) that rejects zero values.
2. **Fetch chain snapshot** filtered to strikes within +/- 40% of
   spot and expiries in the target horizon window (target +/- ~30d).
3. **Pick the nearest expiry** with both calls and puts available.
4. **Build each candidate structure**: find contracts by ATM or OTM
   percentage, use `day.close` (or `fmv` fallback) as the entry
   price. Structures that lack a priceable leg are skipped.
5. **Compute payoff-at-target** analytically per structure.
6. **Rank** by payoff / capital (for directional/vol) or by the
   hedge value (P&L improvement vs unhedged for hedge structures).

## Endpoints used

- `GET /v2/snapshot/locale/us/markets/stocks/tickers/{ticker}` (spot)
- `GET /v3/snapshot/options/{ticker}` (chain)

## Doesn't handle (yet)

- **Prices are day.close, not live.** Delayed on non-realtime
  entitlements; every render surfaces this caveat.
- **Payoff-at-target assumes underlying at target price AT
  EXPIRATION.** Intra-life value depends on IV, theta, and time to
  expiry — not modeled.
- **Greeks omitted when the chain snapshot returns empty greeks**
  (chain endpoint doesn't populate greeks on all keys/tiers). No
  delta/vega/theta context in the current output.
- **Assignment risk and dividend risk on short legs** are ignored.
  Real selection between (say) a bull-call spread and a bull-put
  spread must factor these in outside the tool.
- **Multi-underlying spreads** (calendars, diagonals across expiries)
  are not enumerated. Same-expiry structures only.

Files in this skill

  • README.md2.6 KB
  • SKILL.md4.2 KB
  • output-schema.json3.4 KB
  • references/rendering.md1.7 KB
  • requires.yml1.2 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.